Can diaphragmatic contractility be assessed by twitch airway pressures in patients with chronic obstructive pulmonary disease?

In healthy subjects and in patients without lung diseases, twitch airway pressure (Paw(tw)) responses to phrenic nerve stimulation can be used to predict twitch esophageal pressure (Pes(tw)) and twitch transdiaphragmatic pressure (Pdi(tw)), thus overcoming the need for placement of esophageal and gastric balloons. The aim of this study was to determine whether measurements of Paw(tw) combined with simple maneuvers could be used to predict Pes(tw), and possibly Pdi(tw), in patients with severe chronic obstructive pulmonary disease (COPD) (n = 12). Stimulations delivered at relaxed FRC produced a correlation coefficient (r) between Paw(tw) and Pes(tw) of 0.44 (p < 0.001) and of 0.62 (p < 0.001) during stimulations while patients performed a gentle exhalation from FRC. Stimulations performed during a gentle inhalation produced a good correlation between Paw(tw) and Pes(tw) (r = 0.92, p < 0.001); however, the limits of agreement between Paw(tw) and Pes(tw) were wide. Correlations between Paw(tw) and Pdi(tw) during the three experimental conditions were weak. In conclusion, during a gentle inspiratory effort in patients with severe COPD the correlation between Paw(tw) and Pdi(tw) was weak, whereas the correlation between Paw(tw) and Pes(tw) was good, but it was not sufficient to allow the prediction of Pes(tw) from Paw(tw) in all patients.